A hybrid approach to international market selection: The case of impact investing organizations

International Business Review xxx (xxxx) xxxx

Contents lists available at ScienceDirect

International Business Review journal homepage: www.elsevier.com/locate/ibusrev

A hybrid approach to international market selection: The case of impact investing organizations ⁎

Roy Merslanda, Samuel Anokye Nyarkob, , Amila Buddhika Sirisenac a

University of Agder, School of Business and Law, Universitetsveien 19, 4604 Kristiansand, Norway University of Agder, School of Business and Law and Université Libre de Bruxelles (ULB), SBS-EM, CEB, and CERMi, Avenue F. D. Roosevelt 50, 1050 Brussels, Belgium c University of Agder, School of Business and Law, Universitetsveien 19, 4604 Kristiansand, Norway b

A R T I C LE I N FO

A B S T R A C T

Keywords: Cross-border investments Internationalization Social enterprises International market selection Macroeconomic factors Hybrid organizations

Social enterprises are hybrid organizations that concurrently pursue social and economic goals and hence are mid-way between conventional capitalistic firms and non-profit organizations. Many social enterprises are becoming international; delivering services across borders. With the objective of understanding the internationalization of these unconventional organizations, this paper examines their international market selection decision based on host countries’ macroeconomic conditions. Generally, we hypothesize that the international market selection decision of social enterprises is tied to their hybridity, an overarching characteristic that sets them apart from other types of organizations. We build an original dataset with information on 41 European and North American impact investing organizations and 153 developing countries. Largely, our findings support the hypothesis, suggesting that social enterprises operate in foreign countries that offer a desirable balance between their social and financial goals. However, they avoid contexts with high country risk, factors that could cause a shortfall in expected returns.

JEL classification: F23 G21 L31

1. Introduction ‘The business of doing good,’ or what Miller, Grimes, McMullen, and Vogus (2012, p. 616) term “venturing for others with heart and head,” has become popular. Across the globe, social enterprises are gaining momentum. In the Netherlands, for example, the social enterprise sector grew by more than 70% during the period 2010–2015 (Keizer et al., 2016). Faced with demographic changes and financial crises, governments and the general public have high hopes in social enterprises because these firms promise to address social problems without the need for long-term public (or private) subsidies (Zahra et al., 2009). According to Doherty, Haugh, and Lyon (2014), social enterprises are organizations that strive to achieve desirable social goals, e.g., reducing unemployment, hunger and poverty eradication, while maintaining their financial sustainability. Thus, by pursuing social and financial goals at the same time, social enterprises are hybrid organizations that couple dual institutional logics―social and economic (Battilana & Dorado, 2010). Three characteristics distinguish social enterprises from pure philanthropic organizations and capitalistic firms. The first is their hybridity which stem from the simultaneous pursuit of social and financial

objectives (Battilana & Dorado, 2010). This is perhaps the most overarching and distinct feature of social enterprises as both social and economic value creation is core to them (Peredo & McLean, 2006). Hybridity is also the main source of tension in social enterprises since social and financial logics often conflict (Wry & Zhao, 2018). Second, social enterprises must be financially self-sustainable, implying that they must be able to generate income to cover their costs without donor support (Mair & Marti, 2006; Townsend & Hart, 2008). Social enterprises do so by operating with conventional business models in the delivery of their products and services at the marketplace. Due to their mission orientation and the low economic status of their clients (less privileged people), social enterprises may not charge competitive prices for their products and services. Yet, prices must be high enough to break even at least. This explains why achieving financial sustainability is tricky for most social enterprises (Doherty et al., 2014). Third, social enterprises fill institutional voids that are unattended by governments and the market (Zahra et al., 2009). Thus, social enterprises supply products and services that are unavailable in conventional sectors due to resource constraints faced by governments and other private actors. Such voids are usually costly and unprofitable to fill, a reason for their neglect by the market.

⁎

Corresponding author. E-mail addresses: [email protected] (R. Mersland), [email protected], [email protected] (S.A. Nyarko), [email protected], [email protected] (A.B. Sirisena). https://doi.org/10.1016/j.ibusrev.2019.101624 Received 1 October 2018; Received in revised form 7 August 2019; Accepted 10 September 2019 0969-5931/ © 2019 Elsevier Ltd. All rights reserved.

Please cite this article as: Roy Mersland, Samuel Anokye Nyarko and Amila Buddhika Sirisena, International Business Review, https://doi.org/10.1016/j.ibusrev.2019.101624

International Business Review xxx (xxxx) xxxx

R. Mersland, et al.

The paper proceeds as follows. Section 2 presents the conceptual and theoretical framework. Section 3 outlines the methodological approach and the data while Section 4 presents and discusses the empirical findings. Section 5 presents our conclusions.

In addition to the global popularity of social enterprises, we have in recent years observed a significant increase in cross-border operations by these hybrid firms (Porter & Kramer, 2011). These cross-border activities can be global or regional (McKague et al., 2014; Wang et al., 2015). In some instances, pro-social organizations incorporated in western countries expand their developmental interventions into developing countries either directly or through support-based partnerships with local organizations (Golesorkhi, Mersland, Piekkari, Pishchulov, & Randøy, 2019; Golesorkhi, Mersland, Randøy, & Shenkar, 2019). In most cases, such collaborations involve the transfer of personnel, knowledge and international best practices. Despite the burgeoning literature, the internationalization of social enterprises has received only a paucity of scholarly attention (Pless, 2012; Zahra et al., 2008) and until now, no study has, to the best of our knowledge, investigated the international market selection decisions of social enterprises. We aim to contribute to the literature by explaining the international market decision of social enterprises based on the macroeconomic conditions of host countries. The study also contributes to our understanding of hybrid firms, an understanding which has been a standing call in many previous studies (e.g. See, Battilana et al., 2015; Doherty et al., 2014; Pache & Santos, 2013; Smith et al., 2013). We set out to address the following research question: Into which macroeconomic environment do social enterprises go when investing abroad? We argue that host country macroeconomic conditions have a direct bearing on the ability of hybrid firms to balance the trade-off between their social and financial goals (Ault & Spicer, 2014; Hermes et al., 2011; Smith et al., 2013). For hybrid organizations, the core need to make a social impact distinguishes their internationalization process from those of mainstream firms. Therefore, conventional theories on internationalization may not be sufficient to understand the crossborder operations of social enterprises (Peredo & McLean, 2006). To answer the research question, we use an original dataset comprising of data from 41 impact investing organizations that originate from Europe and North America. Generally, impact investing organizations invest with a dual motive: generating social impact and earning financial returns (Ashta, 2012). By aiming to concurrently achieve both objectives, impact investing firms are faced with trade-offs because these polar goals can conflict (Glac, 2009). For their desired social goal, the impact investing organizations in our dataset contribute to fighting global poverty by providing finance as well as a wide range of nonfinancial assistance to local microfinance institutions (MFIs) in developing countries. MFIs are specialized organizations that are known for alleviating poverty through the provision of banking services to marginalized and disadvantaged persons with income generating activities (Armendáriz & Morduch, 2010). Previous studies have shown that many MFIs rely on their partners in the global North―mostly impact investing organizations―for financing and technical solutions (Mersland et al., 2011; Mersland & Urgeghe, 2013). At the same time, being double bottom line organizations, the impact investors in our dataset equally aim at earning financial returns on their investments in the MFIs. In addition to the data on the impact investing organizations, we also gather macroeconomic data on 153 developing countries. Based on existing literature on hybrid organizations, we generally hypothesize that social enterprises, in our case impact investing organizations, are likely to internationalize into countries where they have the opportunity to balance the competing demands of their dual institutional logics. Thus, social enterprises will target countries that are less developed, institutionally weak, and risky, but not countries where these macroeconomic indicators are at the worst levels. Largely, our empirical investigation supports this hypothesis. In sum, it appears that when going abroad, the average impact investing organization makes an optimum choice by selecting countries that offer a desirable balance in the trade-off between social and economic opportunities. We claim that impact investing organizations adopt this strategy to balance their often conflicting social and financial institutional logics.

2. Conceptual framework: the international market selection of social enterprises We rely on existing literature on hybrid organizations to build a conceptual framework for our empirical work. We acknowledge that hybrid organizations are not restricted to only organizations that blend social and market logics (Pache & Santos, 2013). Nevertheless, existing works have mainly focused on social enterprises. Therefore, we primarily rely on the social enterprise literature to develop our conceptual model and to formulate the research hypotheses. More so, our sample organizations, impact investing organizations, combine same institutional logics―social and business―as other social enterprises do. 2.1. Social enterprises Social enterprises are hybrid firms that fill institutional voids, left unattended by governments and the market, with business-based models (Pache & Santos, 2013; Stevens et al., 2015). Therefore, in regions and societies where government and market failures are commonplace, social enterprises represent important rays of hope (Doherty et al., 2014). A unique characteristic of social enterprises is their hybridity that stems from their subscription to dual institutional logics: social welfare and financial sustainability (Battilana & Dorado, 2010; Battilana & Lee, 2014; Doherty et al., 2014; Pache & Santos, 2013). Being hybrids, social enterprises are neither typical for-profit firms nor typical non-profit firms, but share characteristics of both types of firms (Peredo & McLean, 2006). Social enterprises are often subject to tension in maintaining their hybridity (Battilana et al., 2015; Smith et al., 2013). This tension is a direct consequence of balancing the conflicting demands of the dual institutional logics of social welfare and economic viability (Battilana & Lee, 2014; Doherty et al., 2014). Often, failure to strike a desirable balance between them results in a trade-off, a situation where social enterprises sacrifice the prescriptions and outcomes of one logic in favor of those of the other (Hermes et al., 2011; Jay, 2013; Smith et al., 2013; Wry & Zhao, 2018). Nevertheless, social enterprises endeavour to achieve a satisfactory balance between the two logics since the definition of success encompasses excellence in both logics (Mair & Marti, 2006; Townsend & Hart, 2008). Stated differently, a social enterprise is said to be successful if it attains the feat of creating social value while at the same time being financially self-sustainable (Battilana & Dorado, 2010). On the international scene, social enterprises are confronted with this social-economic tension and need to strike a desirable balance. We demonstrate this using the three host-country macroenvironmental factors—level of development, institutional strength, and country risk―discussed in the next section. We argue that the international market selection decision of social enterprises is largely shaped by their hybridity rather than the prescriptions of conventional approaches. 2.2. International market selection and host-country macroenvironmental climate International market selection is one of the most salient as well as complex decisions an organization has to make during its expansion across borders (Clark et al., 2018; Papadopoulos, Martín Martín, & Gaston‐Breton, 2011). For social enterprises, this decision is highly bounded rational and complex due to inherent operating challenges in developing economies (Papadopoulos & Martín, 2011). Despite the seeming complexity, cross-border activities characterize many hybrid organizations (Zahra et al., 2009). The international market selection of an organization is mainly 2

International Business Review xxx (xxxx) xxxx

R. Mersland, et al.

Hypothesis 2. In selecting international markets, social enterprises target countries with weak institutions but not those with the weakest institutions.

influenced by factors at two levels: target country-level factors and firm-level factors (Kim & Aguilera, 2016). Target country-level factors include market potential, competition, economic factors, political factors, and social factors, while firm-level factors include resources (human, financial, etc.), competencies (technical, managerial, etc.), and organizational goals (Brewer, 2001; Kim & Aguilera, 2016). The present study sheds light on how social enterprises select international markets based on the host-country’s macroeconomic conditions―level of development, institutional strength, and country risk (Bailey, 2017).

2.2.3. Country risk Country risk refers to all factors in a host country that could cause a shortfall in the expected returns from a foreign investment (Meldrum, 2000). This risk is outside the purview of investors and is usually the consequence of imbalances in socio-economic, political, geographic and structural factors between countries (Cosset & Roy, 1991; Meldrum, 2000). Because of country risk, cross border transactions carry incremental risks that are absent in domestic transactions (Meldrum, 2000). In the mainstream management literature, it is theorized that the extent of risk in a target country negatively impacts market selection strategies (Andersen & Buvik, 2002; Brouthers & Nakos, 2005). This is primarily due to the volatile relationship between profitability and risk. Scholars have identified several sources of country risk; e.g., political, social, economic, operational, and transfer and exchange rate risk (Cosset & Roy, 1991; Meldrum, 2000; Root, 1987; Schneider & Frey, 1985). Yet, as far as social enterprises are concerned, the impact of country risk is probably different due to their hybridity. In high-risk countries, vulnerable people and communities are prevalent, thus providing greater opportunity for social enterprises to fulfill their social utility functions (Porter & Kramer, 2011; Teasdale, 2010). At the same time, however, social enterprises need to achieve some level of economic breakthrough in order to advance their social welfare mission. For this reason, high-risk environments may be shaky grounds for social enterprises. Therefore, a country risk level that is unfavourable to the realization of one objective may be favorable to the realization of the other objective, and vice versa (Austin et al., 2006). To balance this trade-off, the optimal choice for social enterprises may be to opt for countries where risk is neither too high nor too low. This brings us to our third hypothesis.

2.2.1. Level of development of host country Market potential is a key determinant of international market selection by traditional firms (Brouthers et al., 2009; Brouthers & Nakos, 2005). Naturally, greater market potential is associated with higher profits, both in present and future terms (Head & Mayer, 2004). To excel, social enterprises require markets with good potential. Although market potential is necessary to guarantee the long-term profitability and growth of social enterprises, it is greater in more developed countries (Hanson, 2005). At the same time, social enterprises have a mandate to tackle diverse societal challenges, such as unemployment, financial and social exclusion, and hunger (Pache & Santos, 2013; Stevens et al., 2015; Townsend & Hart, 2008). These societal challenges and institutional voids are prevalent in most developing countries. As a result, developing countries provide attractive settings and opportunities for social enterprises to create deep social impact (Edwards & Hulme, 1996b). In sum, developed countries offer promising climate to create economic value but less opportunities for creating social value (Edwards & Hulme, 1996b). The reverse is true for poor countries (Edwards & Hulme, 1996a). This is a clear manifestation of the trade-off thesis (Austin et al., 2006; Doherty et al., 2014). Faced with such conflicting demands, social enterprises resort to an optimal choice that balances their social and economic objectives (Mair et al., 2015). Pache and Santos (2013) term this response “selective coupling.” Against this backdrop, we formulate our first hypothesis as follows.

Hypothesis 3. In selecting international markets, social enterprises target countries with high country risk but not those that are most risky.

Hypothesis 1. In selecting international markets, social enterprises target less developed countries but not the least developed ones.

3. Data and methodology

2.2.2. Strength of institutional environment Institutions explain economic growth and the general business environment in a given country, and it has been argued that institutions define the “rules of the game” (North, 1990, p. 3). The purpose of institutions is to protect property rights, enforce contracts between individuals and firms, and provide physical and regulatory infrastructure (Bailey, 2017; North, 1990). Stronger institutions facilitate business transactions and increase the quality of life of individuals by reducing transaction costs (Chen et al., 2018; North, 1990; Roy & Oliver, 2009). Therefore, countries with stronger institutions seem to provide conducive environments for economic exchange (North, 1990; Verbeke & Kano, 2013). This explains why profit-maximizing firms prefer countries with stronger institutions (Chen et al., 2018; Dau, 2013; Murtha & Lenway, 1994). By contrast, countries with weaker institutions are often prone to developmental challenges. In such countries, the by-products of weak institutions, such as corruption, create inequality, deprivation, poverty, poor health care, and various societal ills, are prevalent (Aidt et al., 2008). Because of their social objects, social enterprises regard such developmental challenges stemming from weak institutions as opportunities and the associated countries as natural markets to enter (Koch et al., 2009). On the other hand, these same institutional weaknesses could potentially prevent social enterprises from becoming financially viable, thus posing a threat to their sustainability (Fowler, 1996). Thus, we posit that social enterprises target countries that are positioned somewhere in between, i.e., countries that offer social enterprises the opportunities to earn sufficient profits to pursue social goals. This leads to our second hypothesis.

3.1. Context The present study focuses on European and North American impact investing organizations that operate in developing countries.1 These organizations are incorporated as non-governmental (NGOs), get their income from the services they render rather than from donations, and mainly work in developing countries to promote financial and social inclusion through partnership with local MFIs (Salamon & Anheier, 1992). The microfinance industry offers a natural context for this study since most industry players satisfy the principal criterion for defining a social enterprise, namely, the coupling of social and business logics (Battilana & Dorado, 2010; Peredo & McLean, 2006). Moreover, the microfinance industry is globally known and acknowledged for its commitment to developmental issues. For instance, the United Nations declared 2005 as the year of microcredit and the 2006 Nobel Peace Prize was awarded to microfinance pioneer Mohammad Yunus who founded the Grameen Bank in Bangladesh (one of the first MFIs). Finally, microfinance is a very internationalized industry where international lenders, donors, investors, and technical assistance providers offer their services (Mersland et al., 2011; Mersland & Urgeghe, 2013). Principally, the increasing internationalization of microfinance is 1 The sample of developing countries in the dataset are those classified by the World Bank as upper middle-income, lower middle-income and low-income countries. High-income countries are excluded since they fall outside the mandate of our sample firms.

3

International Business Review xxx (xxxx) xxxx

R. Mersland, et al.

Private Government

Donors

Foundations Investors

International impact investing organizations

Individual Microentrepreneurs

Countrybased MFIs

Reinvested surplus Fig. 1. Flow of Funds. Fig. 1 illustrates how funds flow from suppliers in developed countries to microentrepreneurs in developing countries. Source: Adapted from European Microfinance Platform (2013).

country-based MFIs. The European Microfinance Platform (e-MFP) has 114 members: 104 are organizations, out of which 64 are non-governmental impact investing organizations that provided information for the 2013 directory. However, not all the 64 organizations serve the purposes of this study and therefore we implement a selection procedure that results in a fine-grained sample of 41 impact investing organizations that provide funding and/or technical assistance to MFIs. The filtering of the organizations was done based on two criteria: international presence and type of intervention. Regarding the international presence criterion, only organizations that listed activities in at least one foreign country were selected, leading to the exclusion of 2 organizations that operate solely in their country of origin. For the type of intervention criterion, 4 universities, the United Nations, and 6 oversight organizations were excluded to further align the data with our research interest in double bottom line impact investing organizations. The e-MFP directory’s information was verified from the websites of the respective organizations. In cases of discrepancies and missing information, the organizations were contacted by e-mail for clarifications. After all these, data relating to the operating locations of 11 organization were still missing. After excluding these 11 organizations, the final sample consists of 41

largely driven by an infusion of international funds (Mersland & Urgeghe, 2013). The microfinance industry is thus a suitable testing ground for analyzing patterns of international market selection by social enterprises. 3.2. Sample and data sources The dataset was created by us with data from multiple sources. Our sample of social enterprises consists of impact investing organizations listed in the 2013 directory of the European Microfinance Platform (eMFP). These impact investing organizations, also called microfinance investment vehicles (Mersland & Urgeghe, 2013), channel funds from suppliers (donors and other fund providers) to country-based MFIs, with the aim of achieving mutually beneficial goals (Mersland & Urgeghe, 2013). The relationship between providers of funds in the global north and recipients of credit from MFIs is illustrated in Fig. 1. Besides financial resources, microfinance investment vehicles, especially those incorporated as non-governmental organizations, often provide other non-financial support to their partner MFIs (Mersland et al., 2011). Fig. 2 illustrates the financial and non-financial assistance offered by impact investing organizations to their local partners, the

Subsidies in Money

Financial assistance

Guarantees Loans Equity

International impact investing organizations

Technical Assistance /Capacity Building

Non-financial assistance

Research/Information Dissemination Policy advice/Lobby

Networking/Donor Coordination Fig. 2. Support activities provided International impact investing organizations to country-based MFIs. Fig. 2 illustrates the financial and non-financial assistance that are offered by International impact investing organizations to locally MFIs in developing countries. Source: Adapted from European Microfinance Platform (2013). 4

International Business Review xxx (xxxx) xxxx

R. Mersland, et al.

impact investing organizations2 offering financial and/or non-financial assistance to MFIs in at least one foreign country3 . Country-level macroeconomic data were collected from public sources mentioned in the subsections below.

3.5. Control variables We control for the effects of nine factors: organizational experience, organizational size, distance between home and host countries, type of intervention, bilateral relations between home and host country, bilateral trade, size of host country, religion and host country’s natural resource endowment.

3.3. Dependent variable The dependent variable is a binary variable that indicates whether a given organization operates in a given country: the impact investing organization takes a value of 1 if it operates in the country and 0 otherwise (Coeurderoy & Murray, 2008; Koch et al., 2009).

3.5.1. Experience This is regarded as one of the most important factors in internationalization literature (Davidson, 1980; Kim & Aguilera, 2015). Experienced firms have a better understanding of, and ability to predict, market conditions, thereby reducing their risk and uncertainty (Davidson, 1980). Experience is operationalized by two variables. The first is the age of the organization and the second is international experience which is measured by the total number of countries in which a given organization operates (Dowell & Killaly, 2009; Lu et al., 2014; Mersland et al., 2011).

3.4. Independent variables For the independent variables, three commonly used macroeconomic factors that explain the internationalization of firms are employed: level of development, institutional strength, and country risk.

3.5.2. Firm size A firm’s size, reflected in the amount of resources it controls, plays an important role in formulating its international marketing strategy (Dass, 2000). From a resource-based theory perspective, large firms are able to harness and deploy the required resources that guarantee their internationalization success in a more effective and efficient way than small firms do (Canabal & White, 2008;). In this study, size is measured by the total number of employees in the organization (Dang et al., 2018).

3.4.1. Level of development The Human Development Index (HDI), developed by the United Nations Development Programme, is employed as a proxy for a country’s level of development. According to the United Nations Development Programme, HDI is a compound index that measures a country’s standing in three basic aspects of human development, namely, long and healthy life, schooling, and decent standard of living. Several internationalization studies have approximated the overall level of development of countries based on the HDI (Dow, 2000; Globerman & Shapiro, 2003). To avoid biases resulting from specific year effects, we use the average HDI for the following years: 2000, 2004, 2008, and 2012.

3.5.3. Type of intervention Studies in the internationalization literature have shown that specific firm characteristics―such as product and service offerings, technology, and management attributes―influence the internationalization decisions and processes of firms (e.g., Li, 2018; Ramón-Llorens et al., 2017). Thus, two additional binary variables are included to control for the effects of type of intervention: the provision of financial assistance and non-financial assistance.

3.4.2. Institutional strength This is proxied by the rule of law score, as published by the World Bank (Du et al., 2008; Globerman & Shapiro, 2003). The rule of law “captures perceptions of the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence” (Kaufmann et al., 2011, p. 4). Rule of law scores range from -2.5 to +2.5, representing lower to higher perceptions, respectively. As with the HDI, the average rule of law score for the years 2000, 2004, 2008, and 2012 is used.

3.5.4. Geographical distance Long distance discourages trade between two countries (Dow, 2000; Malhotra et al., 2009). Intuitively, firms more easily extend their operations to neighboring countries than to distant ones (Dow, 2000). Moreover, using data on international alliances in the microfinance industry, Golesorkhi et al., 2019b) report a clear negative relationship between geographical distance between international partners and the MFI’s performance. Geographical distance is operationalized by the direct distance between the capital of the home country and the capital of the host country. Distance data were obtained from two websites that provide distance data between countries: Date and Time (2014) and Geo Bytes (2006).

3.4.3. Country risk This is the risk emanating from socio-political, economic and structural factors in a host country that adversely affects the expected returns or the value of a cross-border investment (Meldrum, 2000). The Euler Hermes Risk Index (EHRI) (Euler Hermes, 2014) is the proxy for country risk (Moser et al., 2008). The EHRI combines five dimensions in determining country risk, including macroeconomic status of the economy, structural soundness of the business environment, political environment, financial flows and cyclical risk indications (Euler Hermes, 2014). The index has values that range from 1 to 4, where higher values represent higher country risks and vice versa. For this indicator, only data relating to the year 2014 are available.

3.5.5. Bilateral relations The flow of investment and social services (such as aid) from developed to developing countries is much influenced by bilateral relations and political arrangements―for example, bilateral investment treaties (Neumayer & Spess, 2005). Accordingly, two controls are included to account for the effects of bilateral relationships between the home and host countries. These include; colonial ties and voting patterns at the United Nations (Neumayer & Spess, 2005; Weiler et al., 2018).

2 The 41 organizations in the sample are headquartered in the following 17 European and North American countries: Italy, Luxembourg, Germany, Spain, Belgium, Ireland, Netherlands, Sweden, Monaco, France, Norway, Switzerland, Denmark, United Kingdom, Liechtenstein, Canada and United States of America. 3 List of all 41 impact investing organizations, their years of establishment, countries of origin, type of intervention, and current international markets are available upon request.

3.5.6. Bilateral trade Charity flows may follow patterns of existing economic ties between countries (Berthélemy & Tichit, 2004; Younas, 2008; Maizels & Nissanke, 1984; Nowak-Lehmann et al., 2009). We account for this in our estimations by controlling for bilateral trade between home and 5

International Business Review xxx (xxxx) xxxx

R. Mersland, et al.

host countries. We use the volume of annual exports from home to host countries as a meaningful proxy (Berthélemy & Tichit, 2004; Metzger et al., 2010). This data is obtained from the database of the International Trade Centre (http://www.intracen.org/).

Pr(Operate ) = β0 + β1 Development + β2 Institution + β3 Country risk + β4 Age + β5 Int . experience + β6 Firm size + β7 Fin. assistance + β8 Nonfin. assistance + β9 Distance + β10 Colony + β11 UN voting

3.5.7. Host country size The proxy for this control is the total population of the respective countries, contained in the Central Intelligence Agency’s World Fact Book (2014). Scholars have argued that populous countries attract more foreign investments thanks to their greater market potential (Nielsen et al., 2017).

+ β12 Export from home + β13 Country size + β14 Christianity + β15 Oil /Gas exporter + ε To capture the non-linear relationship implied by the hypotheses, we test and run the model on the full sample and three sub-samples. The sub-samples are based on the World Bank’s income classification of countries. The first sample, upper middle-income (UMI), are countries with gross national income (GNI) between $4086 and $12,615. The second sample consist of lower middle-income (LMI) countries with GNI values ranging from $1036 to $4085 and the third sample consist of low-income (LI) countries with GNI values lesser than $1035. Descriptive statistics of each sample are reported in Table 3.

3.5.8. Religion Many organizations involved in microfinance are motivated by Christian faith (Mersland et al., 2013). Hence, following Alesina, and Dollar (2000) and Clist (2011), we control for the effects of religion in our models. Religion data is obtained from the Central Intelligence Agency’s World Fact Book.

4. Empirical findings and discussion

3.5.9. Natural resource endowment Social interventions and aid to developing countries may be driven by the selfish interests of donors rather than the needs of recipient countries. These interests may include the quest to gain access and to exploit resources in recipient countries or what Naim (2007) calls ‘rogue aid’. To control for this possible effect, we include a binary variable that indicates whether a host country is an oil and gas exporter (Alesina & Dollar, 2000; Clist, 2011). Oil and gas data is obtained from the Central Intelligence Agency’s World Fact Book. The definitions and summary statistics of all variables are reported in Table 1.

4.1. Descriptive statistics Table 1 presents the descriptive statistics of the variables. The average impact investing organization in the dataset operates in 14.8% of the total sample of countries, corresponding to an approximate number of 23 countries per impact investing organization. The mean level of country development corresponds to an HDI score of 0.580. The mean institutional strength, measured by the World Bank’s Rule of Law index, is negative (-0.432). Thus, most of the countries in the dataset are characterized by weaker institutions. Similarly, the average country risk of 3.159 is high as it gets closer to the maximum possible value of 4. On average, an impact investing organization in the dataset is 28 years old and has about 42 employees. The share of impact investing organizations that offer financial and non-financial services are 73.2% and 95.1% respectively with most organizations combining both interventions. The average distance between the home and host countries is 7042 km. 5.8% of the total sample of developing countries were previous colonies of the countries from which the impact investing organizations originate. Regarding voting patterns at the United Nations, averagely, the countries of origin of the impact investing organizations and the 153 developing countries vote in the same direction in 58.2% of

3.6. Econometric models First, we conduct a two-sample t-test to compare macroeconomic conditions in countries where impact investing organizations operate and countries where they have no operations. We perform this test on four samples (full sample, upper middle-income countries, lower middle-income countries, and low-income countries) to assess if there are univariate differences. The reason why we also run the test on the sub-samples is to better identify the hybridity proposed in the hypotheses. Then, we proceed to a multivariate setting where we specify a probit regression model as follows: Table 1 Definition of variables and descriptive statistics. Variable Dependent variable Operate Independent variables Development Institution Country risk Control variables Age Int. experience Org. size Fin. Assistance Nonfin assistance Distance (ln)Distance Colony UN voting Export from home (ln)Export from home Country Size (ln)Country Size Christianity Oil/Gas exporter

Definition

Obs

Mean

Std. Dev.

Min

Max

“1” if the impact investing organization operates in a given country, “0” otherwise

6,273

0.148

0.355

0

1

Country’s Human Development Index score Country’s score on the World Bank’s measure of rule of law Country’s score on the Euler Hermes Risk Index

5,453 6,027 5,658

0.580 −0.432 3.159

0.152 0.749 1.105

0.270 −2.450 1.000

0.880 1.720 4.000

Age of organization Number of developing countries in which organization operates Number of Employees “1” if the impact investing organization offers financial assistance and “0” otherwise “1” if the impact investing organization offers non-financial assistance and “0” otherwise Geographical distance (in km) between the home and host countries Logarithm of geographical distance between the home and host countries “1” if host country was a colony of home country and “0” otherwise Percentage of agreement between home and host country during voting at the United Nations Volume of export from home to host country (in US$ million) Logarithm of volume of export from home to host country Total population of country (in millions) Logarithm of total population of country “1” if Christianity is the main religion in the host country and “0” otherwise “1” if the host country is an oil and/or gas exporter “0” otherwise

6,273 6,273 6,120 6,273 6,273 6,273 6,273 6,273 5,735 6,035 6,035 6,273 6,273 6,144 6,191

28.196 22.682 41.644 0.732 0.951 7042.382 8.670 0.058 0.582 761.387 10.619 37.50 15.244 0.609 0.272

20.124 18.871 136.995 0.443 0.215 3885.378 0.690 0.234 0.147 4750.376 3.253 150.000 2.429 0.488 0.445

1.000 2 1 0 0 157 5.056 0 0.014 0 0 0.00986 9.196 0 0

72.000 99 874 1 1 49446 10.809 1 1 240000.000 19.297 1350.00 21.024 1 1

6

International Business Review xxx (xxxx) xxxx

R. Mersland, et al.

Table 2 Correlations and Variance Inflation factor.

Development Institution Country risk Age Int. experience Org. size Fin. Assistance Nonfin. assistance Distance Colony UN voting (ln)Export from home (ln)Country Size Christianity Oil/Gas exporter

UN voting (ln)Export from home (ln)Country Size Christianity Oil/Gas exporter

No.

1

2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1.000 0.500 −0.324 0.003 −0.023 0.000 0.003 −0.002 −0.199 −0.080 0.330 0.182 −0.280 0.085 0.118

1.000 −0.444 0.003 −0.031 0.002 0.003 −0.002 −0.093 −0.009 0.289 0.068 −0.420 0.138 −0.183

3

1.000 −0.001 0.011 0.002 0.000 0.002 −0.039 −0.006 −0.150 −0.217 −0.086 −0.213 −0.043

4

5

1.000 0.056 0.267 0.097 0.176 0.028 −0.081 0.097 0.157 −0.003 0.001 −0.001

6

1.000 −0.007 −0.150 0.103 0.000 −0.057 0.031 −0.012 0.029 −0.017 0.004

7

1.000 0.067 0.048 −0.018 −0.019 0.016 0.025 −0.003 −0.001 −0.002

1.000 −0.100 −0.018 0.086 −0.095 0.189 −0.001 0.003 −0.001

8

1.000 0.025 0.041 0.054 0.209 0.001 −0.002 0.000

9

1.000 0.045 −0.389 −0.235 0.000 0.193 0.008

10

VIF

1.000 −0.140 0.017 0.023 0.028 0.008

1.68 2.09 1.59 1.18 1.04 1.08 1.13 1.14 1.38 1.05 1.52 1.57 1.91 1.26 1.17

No.

11

12

13

14

15

11 12 13 14 15

1.000 0.151 −0.032 0.142 −0.130

1.000 0.353 −0.079 0.128

1.000 −0.241 0.180

1.000 −0.095

1.000

cases. The average annual volume of exports from home to host countries is valued at approximately US$ 761.4 million. The mean value of country size, measured by total population, is 37.5 million. Christianity is the main religion in 60.9% of the host countries. For natural resources, 27.2% of the host countries are exporters of oil and/or gas. In Table 2, the correlations between the independent variables are presented. Multicollinearity is a common problem in studies that use macroeconomic data (e.g., Metzger et al., 2010). Multicollinearity is detected when the variance inflation factor of a variable is greater than 5 or when the correlation between two explanatory variables exceeds 0.9 (Hair et al., 2010). Nonetheless, the numbers in Table 2 dispel any concerns of multicollinearity. The correlation coefficients and the variance inflation factor values reported in the table are lower than the aforesaid upper bounds. The highest correlation coefficient is 0.500 (the correlation between development and institution) and the highest variance inflation factor is 2.09. Table 3 gives a brief description of the characteristics of the developing countries in the dataset. A total of 153 developing countries are represented in the dataset. These are countries categorized as upper middle income, lower middle income, or low income by the World Bank4 . Of the 153 countries, 132 host impact investing organizations. The World Bank’s classification of the 132 countries are as follows: 45 are upper middle income, 44 are lower middle income, 34 are lower income, and 9 are unclassified. Naturally, the more developed countries according to the World Bank classification are characterized by higher HDI, better institutions and lower country risk. The one-way ANOVA results reported in the table reveal that the differences observed between the macroeconomic conditions of the respective income categories are statistically significant (p < 0.01). Thus, we show that as one moves from upper middle-income through lower middle-income to low-income countries, the macroeconomic indicators significantly deteriorate, and the countries become more problematic environments for businesses. We rely on this received knowledge to capture the nonlinear relationship implied by the hypotheses and to show the international market selection decisions of impact investing organizations and more generally, that of double bottom line firms. We achieve this

Table 3 Comparison of macroeconomic conditions of countries in the respective income categories using One-Way Analysis of Variance.

Development Institution Country risk

Full sample N = 153

UMI N = 50

LMI N = 46

LI N = 34

F-statistics

0.580 −0.432 3.159

0.700 −0.200 2.766

0.558 −0.541 3.326

0.398 −0.939 3.758

4912.77*** 787.68*** 451.56***

The table shows the characteristics of the sampled countries. There is a total of 153 developing countries in the dataset, classified by the World Bank into upper middle-income (UMI), lower middle-income (LMI), and lower-income (LI) categories.

by performing the analysis on the total sample and the three subsamples of countries as outlined in the methods session. 4.2. Results In the following, we present the main findings of the study. First, we present initial evidence by means of a t-test whereby we compare the microeconomic factors of countries where the impact investing organizations in our sample are present with those of countries where they are absent. Next, we present the probit regression results. 4.3. Mean comparison t-tests and graphical illustration Table 4 presents the mean comparison t-test results. In panel A of Table 4, the test is performed on the full sample of developing countries in the dataset. In panels B and C, the comparison is performed on the sample consisting only of countries in the upper middle-income and lower middle-income categories, respectively. Panel D shows the results of the comparison among countries in the low-income bracket. In panel A, the results show that impact investing organizations generally operate in countries that are significantly less developed and institutionally weaker than the countries where they do not operate. The opposite, however, holds true for the country risk indicator. This finding also holds true in panel B where we consider only upper middleincome countries. In panel C, the mean value of development of countries where impact investing organizations are present is higher than that of countries where they are absent but the difference in means

4 Of the 153 countries, 50 are upper middle-income, 46 are lower middleincome and 34 are low-income A total of 23 countries are not classified by the World Bank into any of the income brackets. Impact investing organizations are present in 9 of these unclassified countries

7

International Business Review xxx (xxxx) xxxx

R. Mersland, et al.

4.4. Multiple regressions

formulated hypotheses. Social enterprises internationalize into poor and institutionally weak countries but avoid the most problematic countries5 . At the same time social enterprises always avoid risky countries. This later finding is counter to our hypothesis. In the full sample of developing countries (1), the coefficient of development is negative and significant (p < 0.01). Thus, impact investing organizations are more likely to target and operate in developing countries that are characterized by low levels of development. In the sample consisting of upper middle-income countries (2), the coefficient of development changes to positive but insignificant. It appears that level of development is not a priority for impact investing organizations in upper middle-income countries. In the sample of lower middle-income countries (3), the development variable has a positive significant coefficient (p < 0.05). In this income category of countries, impact investing organizations prefer to operate in countries with good development. Similarly, in the sample consisting of low-income countries (3), the coefficient of development is positive and significant (p < 0.05). Overall, the regressions show that impact investing organizations internationalize into less developed countries but not the least developed countries. This confirms our first hypothesis. Social enterprises prefer to invest in countries where they can create some social value (Edwards & Hulme, 1996b) without hurting their economic viability (Hanson, 2005). The second macroeconomic factor, institution, is significantly and negatively related to impact investing organizations’ decision to operate in the full sample of developing countries (p < 0.05). Therefore, in general terms, social enterprises are drawn to countries with weak institutional environments (Aidt et al., 2008). The same results are obtained, and conclusions drawn, after running the model on the sample of upper middle-income countries. In the remaining samples―lower middle-income and lower-income countries―the sign of the coefficient changes to positive. Thus, in these income categories, impact investing organizations avoid countries with the weakest institutions. However, the observed positive relationship is only significant in the sample of low-income countries (p < 0.01). Thus, in the quest to maintain their economic viability, social enterprises avoid lowincome countries with the weakest institutions (Dau, 2013; Murtha & Lenway, 1994). Again, this result supports the trade-off hypothesis (Hermes et al., 2011; Jay, 2013; Smith et al., 2013; Wry & Zhao, 2018) and confirm our second hypothesis that social enterprises balance their conflicting objectives by generally entering countries with weak institutions but avoiding those countries with the weakest institutions (Mair & Marti, 2006; Pache & Santos, 2013; Townsend & Hart, 2008). The coefficient of the third macroeconomic variable, country risk, is significantly negative in all estimations. This is interesting because it shows that the organizations in our sample always consider country risk as something negative when entering an international market. In essence, the impact investing organizations in our sample behave as conventional firms do when it comes to a host country’s risk (Andersen & Buvik, 2002; Brouthers & Nakos, 2005; Rothaermel et al., 2006). What kind of country risk could these organizations be avoiding? Indeed, most country risk measures are composite indices of multiple risk components. In a further analysis (unreported), we examine the effects of six (6) components of country risk which we obtained from the database of the Economists Intelligence Unit (http://country.eiu.com/

We run a probit regression first on the full sample consisting of all developing countries in the dataset (1); second, on the sample of upper middle-income countries (2); third, on the sample of lower middle-income countries (3); and finally on the sample of low-income countries (4). Table 5 shows the probit regression results of the macroeconomic determinants of the international market selection decisions of impact investing organizations. The results displayed in the table confirm the univariate differences observed in Table 4 and largely support the

5 In an unreported analysis for robustness checks, we employ alternative proxies for each of the macroeconomic factors. Specifically, development is proxied with the gross domestic product per capita retrieved from the World Bank database, institutional strength is proxied with Transparency International’s corruption perception index (CPI), and country risk is proxied with the country risk classification published by the Organization for Economic Cooperation and Development (OECD). Overall, the results are analogous to those reported in the text and hence support the hybridity hypothesis.

Table 4 t-test comparison of macroeconomic of countries where impact investing organizations operate and countries where they do not operate. Variables

Operate = 1

Panel A: Full sample of developing countries Development 0.536 Institution −0.637 Country risk 3.081 Panel B: Upper middle-income countries Development 0.687 Institution −0.384 Country risk 2.317 Panel C: Lower middle-income countries Development 0.561 Institution −0.579 Country risk 3.029 Panel D: Low-income countries Development 0.402 Institution −0.809 Country risk 3.583

Operate = 0

t-value

0.589 −0.395 3.175

9.729*** 9.066*** 2.354**

0.702 −0.175 2.830

2.853*** 4.693*** 6.383***

0.557 −0.533 3.388

−0.494 1.518 6.199***

0.396 −0.982 3.818

−1.314 −6.934*** 7.795***

In this table, we employ two sample t-tests to compare the microeconomic factors of countries where impact investing organizations operate and countries where they do not operate. *, **, and *** show statistical significance at 0.1, 0.05, and 0.01, respectively.

is too small to be statistically significant. Further, the institutions in the countries where impact investing organizations are present are weaker than the institutions in the countries where they are absent, but similar to development, the difference is not statistically significant. The results also show that impact investing organizations invest in lower middleincome countries where country risk is significantly lower. In panel D, the results show that countries in which impact investing organizations operate have stronger institutions but have similar level of development as countries where they are absent. Again, the risk in the countries where impact investing organizations operate is significantly lower. In line with the hybridity hypothesis, it appears that impact investing organizations internationalize into developing countries that are poor and institutionally weak but keep away from the poorest countries and those with the weakest institutions. The results also suggest that impact investing organizations always avoid high-risk countries. We posit that impact investing organizations approach their international market selection decisions in this way in order to simultaneously “do social good” and be financially self-sustainable. Is there a tipping or turning point in economic conditions where impact investing organizations are most likely to invest? To answer this, we fit a quadratic plot to each of the macroeconomic conditions and the operating tendencies of impact investing organizations. Graphs A, B, and C of Figure 4 show the quadratic plots for level of development, institutional strength, and country risk, respectively. Fig. 3 shows clearly the tipping point of each of the macroeconomic factors. In graph A, the tipping point corresponds to an HDI score of 0.436; in graph B, the tipping point corresponds to a rule of law index of -0.929; and in graph C, the tipping point maps to a Euler Hermes risk index of 2.206. Thus, above or below these points, the probability of investment diminishes.

8

International Business Review xxx (xxxx) xxxx

R. Mersland, et al.

Fig. 3. Quadratic plot of macroeconomic conditions and operating tendencies. Fig. 3 illustrates quadratic plots for each of the macroeconomic factors. Graphs A, B, and C are the plots for level of development, institutional strength, and country risk, respectively.

AllCountries.aspx). These include financial risk (e.g. devaluation risk, marketable debt), foreign trade payments risk (e.g. discriminatory tariffs, trade embargo risk), infrastructure risk (e.g. port facilities, transportation and communication network), macroeconomic risk (e.g. exchange rate volatility, recession risk), political stability risk (e.g. social unrest, orderly transfers) and security risk (e.g. armed conflict, violent crime). Results of this supplementary analysis closely match the main results. It appears that the impact investing organizations in our dataset avoid country risk, regardless of the source. However, we conjecture that this is probably because these organizations are involved in financial intermediation. All the same, the third hypothesis is only partly supported by this result. Some of the control variables yield interesting, significant results. For example, the size of the host country seems to matter when impact investing organizations go global. Populous countries are preferred, as the country size variable is significant in all estimations. This corroborates many other extant studies on mainstream firms (Brouthers et al., 2009; Brouthers & Nakos, 2005). The international experience variable is significantly positive in all regressions, suggesting that the decision to operate in a given county is influenced by the past internationalization experience of the organizations (Davidson, 1980; Kim & Aguilera, 2015). The coefficient of age is mostly positive but significant in models (2) and (3). Intuitively, experienced organizations are more knowledgeable than inexperienced ones. Hence the findings on international experience and age concurs with existing studies that theorize the internationalization process as a function of organizations’ knowledge and their internationalization experience (Johanson & Wiedersheim‐Paul, 1975; Johanson & Vahlne, 1977). The results also

show that impact investing organizations are influenced by bilateral relations between countries when selecting their foreign markets (Neumayer & Spess, 2005; Weiler et al., 2018). This is evidenced by the high significance of Colony in all estimations. Surprisingly, the effect of UN voting is contrary to our expectation as it is negative in all regressions, though significant only in (2) and (3). Perhaps, impact investing organizations’ decision to invest in a country is influenced by the need in the host country as well other forms of bilateral relations (e.g., colonial ties) rather than mere commonalities during UN voting. The finding on geographical distance is particularly interesting. Impact investing organizations do not seem to bother about distance when deciding where to invest. This is contrary to the preference of mainstream firms, which tend to opt for shorter distances when going international (Dow, 2000; Malhotra et al., 2009). A possible explanation for this is that countries classified as developing are far away from Europe and North America; thus, whether a social enterprise enters Uganda or Bolivia does not matter. In any case, it is far away from home (Golesorkhi et al., 2019b). The effect of religion is significantly positive in model (1), suggesting that the organizations in our sample generally invest in countries where Christianity is the main religion. This finding is expected because Christianity is the major religion in most European and Northern American countries, where the impact investing organizations originate. However, the effect of religions vanishes in the models estimated on the sub-samples. Lastly, oil/gas exporter is significantly negative in all regressions, except in (3). This result is unsurprising since oil exporting countries may be well resourced to combat social challenges than others.

9

International Business Review xxx (xxxx) xxxx

R. Mersland, et al.

the type of services, namely, financial intermediation, that they provide. Overall, the optimal choice for social enterprises seems to be to internationalize into countries that offer a desirable balance between social and economic opportunities. We highlight two practical implications of our findings. First, managers of MFIs in developing countries that wish to attract foreign investors (that originate from the global north) should understand and be aware of their own macroeconomic context. This may be an important step to develop the right strategy to mitigate macro-environmental risk. For example, MFIs that operate in weaker economies could attract foreign investors through their commitment to financial sustainability, by showing good social outcomes or by promising higher returns (Cobb et al., 2016). Second, foreign investors should endeavour to look beyond factors at the macro level by considering firm level risks whenever possible. By doing so, foreign investors can assess whether conditions at the firm level compensates for those at the macro-level. Our study contributes to the nascent literature on the internationalization of social enterprises and more generally to the literature on hybrid organizations. The study identifies the host-country macroeconomic factors that social enterprises consider important in their international market selection decisions. In particular, our study sheds light on the hybridity approach that social enterprises adopt in selecting their international markets. Our study provides empirical evidence that social enterprises target foreign markets that enable them to balance their dual institutional logics and thus preserve their hybridity. Based on these findings future studies are encouraged to be mindful of the hybridity of social enterprises when theorizing the internationalization of these firms. Moreover, the hybrid approach to internationalization needs further investigation. Do social enterprises cross-subsidize between countries with good macroeconomic outlook and those with inferior macroeconomic conditions? Do social enterprises initially enter strong economies before weaker ones or vice versa? These possible nuances could be fruitful avenues for future research. Evidence on the specific organizational characteristics of social enterprises that motivate their internationalization could shed further light on the discussion (Brewer, 2001; Nielsen et al., 2017). An example is the role of knowledge (Johanson & Vahlne, 1977). Even though we infer the effect of knowledge through our measures of experience, this approach does not exhaustively capture the role of knowledge (e.g. in mitigating risk) during the internationalization process of social enterprises. Finally, in this paper, we use data from only organizations involved in financial intermediation. Financial institutions that provide credit facilities are concerned about the repayment capacities of their investees. Consequently, such institutions may avoid organizations which operate in countries that have high chances of default. This might be related to why the impact investing organizations in our dataset strongly avoid risky countries regardless of income category. Additionally, the universe of social enterprises is complex and diverse, involving a wide range of players with heterogenous social interventions (Defourny & Nyssens, 2010; Young & Lecy, 2014). Consequently, future studies on other types of social enterprises, i.e., those not involved in financial intermediation, and their internationalization strategies are needed.

Table 5 International market selection and host-country macroeconomic condition. SAMPLE

UMI, LMI & LI UMI LMI Dependent variable: Operate; 1=Yes, 0=No

LI

VARIABLES

(1)

(2)

(3)

(4)

Development

−1.235*** (0.207) −0.111* (0.060) −0.117*** (0.031) 0.002 (0.002) 0.026*** (0.002) 0.000 (0.000) −0.023 (0.080) 0.015 (0.199) 0.091* (0.052) 0.453*** (0.102) −0.406 (0.279) −0.007 (0.012) 0.244*** (0.020) 0.184*** (0.056) −0.245*** (0.059) −5.194*** (0.641) Yes 4,736 0.223 958.4 0.000

1.049 (0.739) −1.050*** (0.265) −0.416*** (0.112) 0.008** (0.004) 0.032*** (0.004) 0.000 (0.001) −0.052 (0.178) −0.292 (0.421) 0.125 (0.105) 0.502** (0.252) −1.519* (0.880) 0.006 (0.029) 0.245*** (0.065) −0.094 (0.163) −0.330* (0.186) −6.272*** (1.664) Yes 1,635 0.348 340.9 0.000

1.207** (0.485) 0.067 (0.105) −0.119** (0.046) 0.006** (0.003) 0.029*** (0.003) 0.000 (0.000) −0.271** (0.136) −0.116 (0.388) 0.075 (0.089) 0.321* (0.174) −0.905** (0.429) 0.038* (0.022) 0.160*** (0.033) 0.136 (0.110) −0.076 (0.097) −5.119*** (1.070) Yes 1,517 0.207 306.9 0.000

1.432** (0.572) 0.444*** (0.131) −0.536*** (0.121) −0.004 (0.003) 0.029*** (0.003) −0.001 (0.000) 0.238* (0.140) −0.054 (0.315) −0.206 (0.177) 0.723*** (0.174) −0.907 (0.650) −0.018 (0.022) 0.254*** (0.049) 0.085 (0.095) −0.772*** (0.249) −1.353 (1.654) Yes 1,265 0.246 361.7 0.000

Institution Country risk Age Int. experience Org. size Fin. Assistance Nonfin. assistance Distance Colony UN voting (ln)Export from home Country size Christianity Oil/Gas exporter Constant Origin dummies Observations Pseudo R2 LR χ2 Prob > χ2

This table shows the probit regression results of the macroeconomic determinants of the international market selection decisions of impact investing organizations. UMI = upper middle income, LMI = lower middle income, and LI = low income. Observations = product of total number of impact investing organizations and number of countries in the sample. Standard errors are in parentheses. *, **, and *** show statistical significance at 0.1, 0.05, and 0.01, respectively.

5. Conclusions In this article, the international market selection of social enterprises is examined based on the macroeconomic conditions of the host countries. By investigating this relationship, our aim is to shed light on the location preferences of social enterprises, in terms of macroeconomic conditions in the host country, when they go international and whether this is tied to their hybridity. This phenomenon is explored using data from 41 impact investing organizations that on average operate in 23 developing counties. The empirical results reveal that impact investing organizations that expand their activities across borders target less developed and institutionally weak countries. However, they do not target the least developed countries and those with the weakest institutions. We argue that this is because social enterprises must balance their social and financial logics (Mair & Marti, 2006; Mair et al., 2015; Pache & Santos, 2013). Thus, social enterprises fulfill their social obligation by targeting poorly developed countries with weak institutions, but at the same time they ensure their financial sustainability by not entering the most problematic countries. The study further shows that impact investing organizations avoid high-risk countries, a finding that may be related to

Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Declaration of Competing Interest None Acknowledgements We are grateful to the editor Pervez Ghauri, Jorma Larimo, Ariane 10

International Business Review xxx (xxxx) xxxx

R. Mersland, et al.

Szafarz, Arvind Ashta, François-Xavier Ledru and three anonymous reviewers for their useful comments and suggestions. We also thank the participants of the following conferences for their valuable comments and suggestions: 7th Aalborg International Business Conference (Aalborg. May 2018), the 2018 Center for Research on Social Enterprises and Microfinance (CERSEM) Research Day (Kristiansand. April 2018), the Centre for European Research in Microfinance (CERMi) Research Day (Brussels. October 2018), 44th European International Business Academy Conference (Poznań. 2018) and the 6th European Research Conference on Microfinance (Paris. June 2019).

417–436. Dow, D. (2000). A note on psychological distance and export market selection. Journal of International Marketing, 8(1), 51–64. Dowell, G., & Killaly, B. (2009). Effect of resource variation and firm experience on market entry decisions: Evidence from US telecommunication firms’ international expansion decisions. Organization Science, 20(1), 69–84. Du, J., Lu, Y., & Tao, Z. (2008). Economic institutions and FDI location choice: Evidence from US multinationals in China. Journal of Comparative Economics, 36(3), 412–429. Edwards, M., & Hulme, D. (1996a). Beyond the magic bullet: NGO performance and accountability in the post-cold war world. West Hartford, CT: Kumarian Press. Edwards, M., & Hulme, D. (1996b). Too close for comfort? The impact of official aid on nongovernmental organizations. World Development, 24(6), 961–973. Hermes, E. (2014). Country riskRetrieved fromhttp://www.eulerhermes.com/economicresearch/country-reports/Pages/default.aspx. Fowler, A. (1996). Demonstrating NGO performance: Problems and possibilities. Development in Practice, 6(1), 58–65. Geo Bytes (2006). City distance tool. Retrieved fromhttp://www.geobytes.com/ CityDistanceTool.htm. Glac, K. (2009). Understanding socially responsible investing: The effect of decision frames and trade-off options. Journal of Business Ethics, 87(1), 41–55. Globerman, S., & Shapiro, D. (2003). Governance infrastructure and US foreign direct investment. Journal of International Business Studies, 34(1), 19–39. Golesorkhi, S., Mersland, R., Piekkari, R., Pishchulov, G., & Randøy, T. (2019a). The effect of language use on the financial performance of microfinance banks: Evidence from cross-border activities in 74 countries. Journal of World Business. https://doi.org/10. 1016/j.jwb.2019.03.002. Golesorkhi, S., Mersland, R., Randøy, T., & Shenkar, O. (2019b). The performance impact of informal and formal institutional differences in cross-border alliances: Insights from the microfinance industry. International Business Review, 28(1), 104–118. Hair, J., Black, W., Babin, B., & Anderson, R. (2010). Multivariate data analysis: A global perspective (7 ed.). Upper Saddle River, NJ: Prentice Hall. Hanson, G. H. (2005). Market potential, increasing returns and geographic concentration. Journal of International Economics, 67(1), 1–24. Head, K., & Mayer, T. (2004). Market potential and the location of Japanese investment in the European Union. The Review of Economics and Statistics, 86(4), 959–972. Hermes, N., Lensink, R., & Meesters, A. (2011). Outreach and efficiency of microfinance institutions. World Development, 39(6), 938–948. Jay, J. (2013). Navigating paradox as a mechanism of change and innovation in hybrid organizations. The Academy of Management Journal, 56(1), 137–159. Johanson, J., & Vahlne, J. E. (1977). The internationalization process of the firm—A model of knowledge development and increasing foreign market commitments. Journal of International Business Studies, 8(1), 23–32. Johanson, J., & Wiedersheim‐Paul, F. (1975). The internationalization of the firm—Four swedish cases. Journal of Management Studies, 12(3), 305–323. Kaufmann, D., Kraay, A., & Mastruzzi, M. (2011). The worldwide governance indicators: A summary of methodology, data and analytical issues. Hague Journal on the Rule of Law, 3(2), 220–246. Keizer, A., Stikkers, A., Heijmans, H., Carsouw, R., & Aanholt, W. V. (2016). Scaling the impact of the social enterprise sector. Retrieved fromhttps://www.mckinsey.com/ industries/social-sector/our-insights/scaling-the-impact-of-the-social-enterprisesector. Kim, J. U., & Aguilera, R. V. (2015). Foreign location choice: Review and extensions. International Journal of Management Reviews, 18(2), 133–159. Kim, J. U., & Aguilera, R. V. (2016). Foreign location choice: Review and extensions. International Journal of Management Reviews, 18(2), 133–159. Koch, D.-J., Dreher, A., Nunnenkamp, P., & Thiele, R. (2009). Keeping a low profile: What determines the allocation of aid by non-governmental organizations? World Development, 37(5), 902–918. Li, P.-Y. (2018). Top management team characteristics and firm internationalization: The moderating role of the size of middle managers. International Business Review, 27(1), 125–138. Lu, J., Liu, X., Wright, M., & Filatotchev, I. (2014). International experience and FDI location choices of Chinese firms: The moderating effects of home country government support and host country institutions. Journal of International Business Studies, 45(4), 428–449. Mair, J., & Marti, I. (2006). Social entrepreneurship research: A source of explanation, prediction, and delight. Journal of World Business, 41(1), 36–44. Mair, J., Mayer, J., & Lutz, E. (2015). Navigating institutional plurality: Organizational governance in hybrid organizations. Organization Studies, 36(6), 713–739. Maizels, A., & Nissanke, M. K. (1984). Motivations for aid to developing countries. World Development, 12(9), 879–900. Malhotra, S., Sivakumar, K., & Zhu, P. (2009). Distance factors and target market selection: The moderating effect of market potential. International Marketing Review, 26(6), 651–673. McKague, K., Menke, M., & Arasaratnam, A. (2014). Access afya: Micro-clinic health franchise designed for scale. In I. Alon (Ed.). Social franchising (pp. 61–79). London: Palgrave Pivot. Meldrum, D. (2000). Country risk and foreign direct investment. Business Economics, 35(1), 33–40. Mersland, R., D’Espallier, B., & Supphellen, M. (2013). The effect of religion on development efforts: Evidence from the microfinance industry and a research agenda. World Development, 41(1), 145–156. Mersland, R., & Urgeghe, L. (2013). International debt financing and performance of microfinance institutions. Strategic Change: Briefings in Entrepreneurial Finance, 22(1–2), 17–29. Mersland, R., Randøy, T., & Strøm, R.Ø. (2011). The impact of international influence on

References Aidt, T., Dutta, J., & Sena, V. (2008). Governance regimes, corruption and growth: Theory and evidence. Journal of Comparative Economics, 36(2), 195–220. Alesina, A., & Dollar, D. (2000). Who gives foreign aid to whom and why? Journal of Economic Growth, 5(1), 33–63. Andersen, O., & Buvik, A. (2002). Firms’ internationalization and alternative approaches to the international customer/market selection. International Business Review, 11(3), 347–363. Armendáriz, B., & Morduch, J. (2010). The economics of microfinance. Cambridge, MA: MIT Press. Ashta, A. (2012). Co-creation for impact investment in microfinance. Strategic Change, 21(1/2), 71–81. Austin, J., Stevenson, H., & Wei‐Skillern, J. (2006). Social and commercial entrepreneurship: Same, different, or both? Entrepreneurship Theory and Practice, 30(1), 1–22. Bailey, N. (2017). Exploring the relationship between institutional factors and FDI attractiveness: A meta-analytic review. International Business Review, 27(1), 139–148. Battilana, J., & Dorado, S. (2010). Building sustainable hybrid organizations: The case of commercial microfinance organizations. The Academy of Management Journal, 53(6), 1419–1440. Battilana, J., & Lee, M. (2014). Advancing research on hybrid organizing: Insights from the study of social enterprises. The Academy of Management Annals, 8(1), 397–441. Battilana, J., Sengul, M., Pache, A.-C., & Model, J. (2015). Harnessing productive tensions in hybrid organizations: The case of work integration social enterprises. The Academy of Management Journal, 58(6), 1658–1685. Berthélemy, J. C., & Tichit, A. (2004). Bilateral donors’ aid allocation decisions—A threedimensional panel analysis. International Review of Economics & Finance, 13(3), 253–274. Brewer, P. (2001). International market selection: Developing a model from Australian case studies. International Business Review, 10(2), 155–174. Brouthers, L. E., & Nakos, G. (2005). The role of systematic international market selection on small firms’ export performance. Journal of Small Business Management, 43(4), 363–381. Brouthers, L. E., Mukhopadhyay, S., Wilkinson, T. J., & Brouthers, K. D. (2009). International market selection and subsidiary performance: A neural network approach. Journal of World Business, 44(3), 262–273. Canabal, A., & White, G. O. (2008). Entry mode research: Past and future. International Business Review, 17(3), 267–284. Chen, J., Saarenketo, S., & Puumalainen, K. (2018). Home country institutions, social value orientation, and the internationalization of ventures. International Business Review, 27(2), 443–454. Clark, D. R., Li, D., & Shepherd, D. A. (2018). Country familiarity in the initial stage of foreign market selection. Journal of International Business Studies, 49(4), 442–472. Clist, P. (2011). 25 years of aid allocation practice: Whither selectivity? World Development, 39(10), 1724–1734. Cobb, J. A., Wry, T., & Zhao, E. Y. (2016). Funding financial inclusion: Institutional logics and the contextual contingency of funding for microfinance organizations. The Academy of Management Journal, 59(6), 2103–2131. Coeurderoy, R., & Murray, G. (2008). Regulatory environments and the location decision: Evidence from the early foreign market entries of new-technology-based firms. Journal of International Business Studies, 39(4), 670–687. Cosset, J. C., & Roy, J. (1991). The determinants of country risk ratings. Journal of International Business Studies, 22(1), 135–142. Dang, C., Li, Z. F., & Yang, C. (2018). Measuring firm size in empirical corporate finance. Journal of Banking & Finance, 86, 159–176. Dass, P. (2000). Relationship of firm size, initial diversification, and internationalization with strategic change. Journal of Business Research, 48(2), 135–146. Date and Time (2014). Distance between cities. Retrieved fromhttp://dateandtime.info/ distance.php?id1=4140963&id2=1248991. Dau, L. A. (2013). Learning across geographic space: Pro-market reforms, multinationalization strategy, and profitability. Journal of International Business Studies, 44(3), 235–262. Davidson, W. H. (1980). The location of foreign direct investment activity: Country characteristics and experience effects. Journal of International Business Studies, 11(2), 9–22. Defourny, J., & Nyssens, M. (2010). Conceptions of social enterprise and social entrepreneurship in Europe and the United States: Convergences and divergences. Journal of Social Entrepreneurship, 1(1) 32–53. Doherty, B., Haugh, H., & Lyon, F. (2014). Social enterprises as hybrid organizations: A review and research agenda. International Journal of Management Reviews, 16,

11

International Business Review xxx (xxxx) xxxx

R. Mersland, et al.

Roy, J.-P., & Oliver, C. (2009). International joint venture partner selection: The role of the host-country legal environment. Journal of International Business Studies, 40(5), 779–801. Salamon, L. M., & Anheier, H. K. (1992). In search of the non-profit sector II: The problem of classification. VOLUNTAS International Journal of Voluntary and Nonprofit Organizations, 3(3), 267–309. Schneider, F., & Frey, B. S. (1985). Economic and political determinants of foreign direct investment. World Development, 13(2), 161–175. Smith, W. K., Gonin, M., & Besharov, M. L. (2013). Managing social-business tensions: A review and research agenda for social enterprise. Business Ethics Quarterly, 23(3), 407–442. Stevens, R., Moray, N., & Bruneel, J. (2015). The social and economic mission of social enterprises: Dimensions, measurement, validation, and relation. Entrepreneurship Theory and Practice, 39(5), 1051–1082. Teasdale, S. (2010). How can social enterprise address disadvantage? Evidence from an inner city community. Journal of Nonprofit & Public Sector Marketing, 22(2), 89–107. Townsend, D. M., & Hart, T. A. (2008). Perceived institutional ambiguity and the choice of organizational form in social entrepreneurial ventures. Entrepreneurship Theory and Practice, 32(4), 685–700. Verbeke, A., & Kano, L. (2013). The transaction cost economics (TCE) theory of trading favors. Asia Pacific Journal of Management, 30(2), 409–431. Wang, H., Alon, I., & Kimble, C. (2015). Dialogue in the dark: Shedding light on the development of SEs in China. Global Business and Organizational Excellence, 34(4), 60–69. Weiler, F., Klöck, C., & Dornan, M. (2018). Vulnerability, good governance, or donor interests? The allocation of aid for climate change adaptation. World Development, 104, 65–77. Wry, T., & Zhao, E. Y. (2018). Taking trade-offs seriously: Examining the contextually contingent relationship between social outreach intensity and financial sustainability in global microfinance. Organization Science doi.org/10.1287/orsc.2017.1188. Younas, J. (2008). Motivation for bilateral aid allocation: Altruism or trade benefits. European Journal of Political Economy, 24(3), 661–674. Young, D. R., & Lecy, J. D. (2014). Defining the universe of social enterprise: Competing metaphors. VOLUNTAS: International Journal of Voluntary and Nonprofit Organisations, 25(5) 1307–1332. Zahra, S. A., Gedajlovic, E., Neubaum, D. O., & Shulman, J. M. (2009). A typology of social entrepreneurs: Motives, search processes and ethical challenges. Journal of Business Venturing, 24(5), 519–532. Zahra, S. A., Rawhouser, H. N., Bhawe, N., Neubaum, D. O., & Hayton, J. C. (2008). Globalization of social entrepreneurship opportunities. Strategic Entrepreneurship Journal, 2(2), 117–131.

microbanks’ performance: A global survey. International Business Review, 20(2), 163–176. Metzger, L., Nunnenkamp, P., & Mahmoud, T. O. (2010). Is corporate aid targeted to poor and deserving countries? A case study of Nestlé’s aid allocation. World Development, 38(3), 228–243. Miller, T. L., Grimes, M. G., McMullen, J. S., & Vogus, T. J. (2012). Venturing for others with heart and head: How compassion encourages social entrepreneurship. The Academy of Management Review, 37(4), 616–640. Moser, C., Nestmann, T., & Wedow, M. (2008). Political risk and export promotion: Evidence from Germany. The World Economy, 31(6), 781–803. Murtha, T. P., & Lenway, S. A. (1994). Country capabilities and the strategic state: How national political institutions affect multinational corporations’ strategies. Strategic Management Journal, 15(S2), 113–129. Naim, M. (2007). Rogue aid. Foreign Policy, 159, 95–96. Neumayer, E., & Spess, L. (2005). Do bilateral investment treaties increase foreign direct investment to developing countries? World Development, 33(10), 1567–1585. Nielsen, B. B., Asmussen, C. G., & Weatherall, C. D. (2017). The location choice of foreign direct investments: Empirical evidence and methodological challenges. Journal of World Business, 52(1), 62–82. North, D. C. (1990). Institutions, institutional change and economic performance. Cambridge: Cambridge University Press. Nowak-Lehmann, D. F., Martínez-Zarzoso, I., Klasen, S., & Herzer, D. (2009). Aid and trade–A donor’s perspective. The Journal of Development Studies, 45(7), 1184–1202. Pache, A.-C., & Santos, F. (2013). Inside the hybrid organization: Selective coupling as a response to competing institutional logics. The Academy of Management Journal, 56(4), 972–1001. Papadopoulos, N., & Martín, O. M. (2011). International market selection and segmentation: Perspectives and challenges. International Marketing Review, 28(2), 132–149. Papadopoulos, N., Martín Martín, O., & Gaston-Breton, C. (2011). International market selection and segmentation: A two‐stage model. International Marketing Review, 28(3), 267–290. Peredo, A. M., & McLean, M. (2006). Social entrepreneurship: A critical review of the concept. Journal of World Business, 41(1), 56–65. Pless, N. M. (2012). Social entrepreneurship in theory and practice: An introduction. Journal of Business Ethics, 111(3), 317–320. Porter, M. E., & Kramer, M. R. (2011). Creating shared value. Harvard Business Review, 89(1), 2–18. Ramón-Llorens, M. C., García-Meca, E., & Duréndez, A. (2017). Influence of CEO characteristics in family firms internationalization. International Business Review, 26(4), 786–799. Root, F. R. (1987). Entry strategies for international market. Lexington, MA: Lexington Books.

12